MicroRNA-10 modulates Hox genes expression during Nile tilapia embryonic
development
Juliana Giustia, Danillo Pinhal b, Simon Moxonc, Camila Lovaglio Camposb, Andrea
Münsterbergd and Cesar Martinsa*
a- Institute of Biosciences, Department of Morphology - Sao Paulo State University, Botucatu,
Brazil. [email protected]; [email protected] Institute of Biosciences, Department of Genetics - Sao Paulo State University, Botucatu,
Brazil. [email protected]; [email protected] The Genome Analysis Centre, Norwich Research Park - Norwich - UK.
[email protected] School of Biological Sciences, University of East Anglia, Norwich Research Park -
Norwich - UK. [email protected]
*Corresponding author: E-mail: [email protected], [email protected]
1
Abstract
Hox gene clusters encode a family of transcription factors that govern anterior-posterior axis
patterning during embryogenesis in all bilaterian animals. The time and place of Hox gene
expression are largely determined by the relative position of each gene within its cluster.
Furthermore, Hox genes were shown to have their expression fine-tuned by regulatory
microRNAs (miRNAs). However, the mechanisms of miRNA-mediated regulation of these
transcription factors during fish early development remain largely unknown. Here we have
profiled three highly expressed miR-10 family members of Nile tilapia at early embryonic
development, determined their genomic organization as well as performed functional
experiments for validation of target genes. Quantitative analysis during developmental stages
showed miR-10 family expression negatively correlates with the expression of HoxA3a,
HoxB3a and HoxD10a genes, as expected for bona fide miRNA-mRNA interactions.
Moreover, luciferase assays demonstrated that HoxB3a and HoxD10a are targeted by miR-
10b-5p. Overall, our data indicate that the miR-10 family directly regulates members of the
Hox gene family during Nile tilapia embryogenesis.
Keywords: Oreochromis niloticus; embryos; HoxA3a; HoxB3a; HoxD10a; miRNA
2
1. Introduction
The development of a multicellular organism from a single cell is a complex process
involving many molecules including transcription factors, which must be present in specific
cells at the right time (Montavon et al., 2011). A group of master transcription factors are
encoded by the Hox gene clusters that have conserved roles in patterning the anterior-
posterior axis during embryogenesis in all bilaterian animals (Alexander et al., 2009). The
mechanisms of transcriptional regulation of these molecules are dictated by the organization
of the genes in clusters within the genome (Duboule et al., 2007). This type of genomic
organization allows for sharing nuclear space, chromatin structure, common regulatory
elements, such as enhancers, and even promoters. Furthermore, it provides time and spatial
colinearities during development, because Hox genes located closer to the enhancers are
transcribed earlier (Andrey et al., 2013). As a result, the time and place of Hox gene
expression are largely determined by the relative position of each gene within its cluster
(Duboule et al., 2007).
Hox genes from the same group (transparalogous or paralogues genes) arose from
duplication and share more similarity in protein sequence and expression pattern than other
genes within a cluster. In mice and other mammals there are 39 Hox genes arranged in four
clusters (A, B, C and D) located on four different chromosomes, whereas teleost fishes have
at least 48 Hox genes in eight clusters (Aa, Ab, Ba, Bb, Ca, Cb, Da and Db) that resulted from
a whole genome duplication (Amores et al., 2004). Intriguingly elasmobranchs (sharks and
rays) have only 3 Hox clusters (A, B and D) as consequence of a genomic deletion of their
entire HoxC cluster (King et al., 2011). The Nile tilapia Oreochromis niloticus has 51 Hox
genes arranged in seven clusters, (Figure 1). Several studies have compared Hox gene
organization, and Hox of Nile tilapia seems to be more similar to orthologues of pufferfish
(Tetraodon nigrovirides) and medaka (Oryzias latipes) than to zebrafish (Santini & Bernardi,
2005).
The observed complexity of Hox genes regarding paralogs diversity and arrangement
patterns in teleost fish is a consequence of three rounds of genome duplications that these
animals are believed to have undergone during vertebrate evolution. Following the first
duplication (500 My), the AB cluster lost the Hox12 gene and the CD cluster lost the Hox2
and Hox7 genes. After the second duplication, the A cluster lost the Hox8 gene, the B cluster
lost the Hox11 gene, the C cluster lost the Evx gene and the D cluster lost the Hox6 gene. The
divergence of the tetrapod lineage followed the second duplication (400 My) and tetrapods
underwent specific gene losses while the ancient ray-finned fish (teleosts) ancestor underwent
the third duplication (350 My) and lost several genes in all the clusters. The organization of
the “a” clusters appears to be very conserved, whereas “b” clusters have lost more genes
(Santini & Bernardi, 2005).
3
Paralogous Hox genes often perform distinct biological roles, as evidenced by their
mutant phenotypes, but may also show extensive redundancy and functional overlap. In fruit
flies and mice, deletion of a single Hox gene leads to altered axial identities and
transformation of specific embryonic structures into more anterior ones (Rijli et al., 1973;
Kaufman et al., 1978; Gendron-Maguire et al., 1993). Conversely, ectopic expression of a
single Hox gene can also result in a posterior transformation or loss of the body structures
(Denell et al., 1981; Van de Ven et al., 2011), thereby interfering permanently with
organismal development.
Development is a complex process requiring several events to be accurately
temporally and spatially regulated. In this sense, microRNAs (miRNAs) were reported as key
regulatory elements for proper organism development based on their ability to modulate gene
expression, including the expression of transcriptional factors (Mallo & Alonso, 2013). As
members of an abundant class of small noncoding RNAs, miRNAs repress gene expression
by preferentially binding to complementary target sequences in the 3'UTRs of mRNAs
leading to mRNA degradation and/or translational repression (Bartel, 2009; Lee & Shin,
2012).
MiRNA-mediated regulation of Hox genes has been previously reported in
Drosophila (Bender, 2008), mouse (Mansfield & McGlinn, 2012), chick (Wong et al., 2015)
and human (Lund, 2010) implying that regulation via miRNAs comprises an extra tier in the
complex molecular regulatory circuit controlling Hox gene expression. For instance,
downregulation of miR-10 in zebrafish embryos leads to overexpression of HoxB1a and
HoxB3a (Woltering & Durston, 2008) and in humans, miR-10 downregulation was negatively
correlated with HoxA1 overexpression (Garzon et al., 2006). These data suggests that the
same miRNA may target paralogous genes from distinct Hox clusters.
Notably, several Hox-regulating miRNAs of vertebrates are encoded within the Hox
clusters, as observed for miR-10 and miR-196. In mammals, miR-10a resides upstream of
HoxB4 and miR-10b is upstream of HoxD4. This intronic genomic arrangement might
provide an effective mechanism for the co-expression of miRNAs and their Hox mRNA
targets in a temporal and spatial manner (Tanzer et al., 2005; Mallo & Alonso, 2013). In
addition, zebrafish and human genomes have intergenic miRNAs encoded outside of the Hox
clusters such as mir-99a, mir-99b and mir-100, highly homologous to mir-10a and mir-10b,
that despite nucleotide difference within seed region, may have overlapping targets (Tehler et
al., 2011; Woltering & Durston, 2008). These data shows that miRNAs encoded/associated
with Hox clusters can differentially modulate the expression of Hox during vertebrate
development. In Nile tilapia, however, the spatiotemporal expression profiles of miR-10
family and their modulation over Hox genes remain poorly investigated.
4
Although significant evidence has been generated regarding the biological roles of
miR-10 family members, further experiments are required to determine the specific genes
they target, which in turn, will reveal the physiological functions regulated by them.
Moreover, given the dynamism of fish genomes, particularly of cichlid species, and the
complex evolutionary history of gene birth and death of Hox clusters in vertebrates, the
correct description of miRNA regulation over these transcription factors in Nile tilapia
requires a detailed inspection of miR-10 family genomic organization.
In this paper, we investigate the role of miR-10 family members in the regulation of a
number of Hox genes (HoxA3a, HoxB3a and HoxD10a) during the ontogenesis of Nile tilapia
fish. For this purpose, we firstly predicted miR-10 family targets through bioinformatics
approaches. Subsequently, we quantified and correlated the expression profiles of both miR-
10 family members and target genes at several embryonic developmental stages. Lastly, we
validated miRNA-target interactions by functional in vitro assays and examined miR-10
genomic organization. Our results demonstrated that HoxB3a and HoxD10a are regulated by
miR-10b-5p during the early development of Nile tilapia, providing input for future research
in vertebrates and in fish development.
2. Materials and Methods
2.1. Samples and RNA purification
All procedures involving animals were performed according to principles set by the
Ethics Committee for Animal Experimentation –Institute of Biosciences – São Paulo State
University (protocol 34/08). All fish were anesthetized with benzocaine (100mg/L of water)
before being euthanized in liquid nitrogen.
Nile tilapia embryos (1, 3, 5 and 7 days post fertilization - dpf) and 30 dpf juveniles
of both sexes were collected in the Royal Fish Farm, Jundiaí, São Paulo, Brazil. Embryos
were removed from the female mouth and selected based on morphology (Fujimura & Okada,
2007) with a stereomicroscope. These sampling periods were based on cell differentiation
stage of the embryonic development cycle of O. niloticus (Fujimura & Okada, 2007; Ijiri et
al., 2008). The 30 dpf period represents sex-differentiated animals. All specimens were placed
in a solution of benzocaine, frozen in liquid nitrogen and stored in -80°C until RNA isolation.
RNA isolation was performed using TRIZOL kit (Invitrogen) according to the
manufacturer's instructions. RNA recovered from samples was quantified by
spectrophotometry (NanoVue, GE Healthcare Life Sciences) and checked for quality through
the RNA integrity number (RIN) analysis. RNA samples were treated with DNA FreeTM Kit
(Ambion) to remove genomic DNA contamination.
2.2. qRT-PCR of miRNAs
5
Expression of five selected miR-10 family members (Table 1) was measured in
embryos at 1, 3, 5 and 7 dpf life stages and in 30 dpf juveniles by qRT-PCR. These miRNAs
were preferentially chosen based on previous small RNA deep sequencing data of our
research group, which verified that they are 6-fold more higly expressed in Nile tilapia
embryos than other miRNAs (Pinhal et. al, unpublished data).
From total RNA, mature miRNAs were converted into cDNA using TaqMan
MicroRNA reverse transcription kit (Life Technologies) following the manufacturer's
instructions. Subsequently, qRT-PCR was carried out using TaqMan 2x Universal Master
Mix 1x, TaqMan MicroRNA Assay Mix 1x, 2ng/uL cDNA and the volume was completed to
20µL with nuclease-free water. In these experiments, the endogenous U6 snRNA was used as
a reference gene.
2.3. Target prediction and quantitative expression analysis
Among several Hox genes known to be in Nile tilapia genome, HoxB3a and
HoxD10a were found to be potentially targeted by miR-10b-5p, miR-10 family members,
based on TargetScan (Grimson et al., 2007), Pictar (Krek et al., 2005) and miRanda (Enright
et al., 2003) prediction tools outputs. These genes were then quantified by qRT-PCR on
samples from 5 animals of both sexes and three experimental replicates for each
developmental period. Reverse transcription of total mRNA was performed using the High
Capacity RNA-to-cDNA Master Mix kit (Life Technologies) according to the manufacturer's
guidelines. qRT-PCR was performed using 1xGoTaq® probe qPCR master mix based on
SYBR Green chemistry (Promega), 40ng/uL of RT reaction, 900nM of primers (forward and
reverse) (Table 1) to 10 mM and the final volume was completed to 20µL with nuclease-free
water.
Thermocycling was performed on a Step-one PCR System (Applied Biosystems) and
reaction conditions were 2 min at 50°C, 10 min at 95°C to polymerase activation, followed by
40 cycles of 15 sec of 95°C and 1 min at 60°C. The relative expression of target genes was
evaluated using the comparative quantification method and the hypoxanthine
phosphoribosyltransferase gene (HPRT) was used as an endogenous control.
2.4. Luciferase reporter constructs
Functional validation of miR-10b-5p action on HoxB3a and HoxD10a genes were
based on the luciferase gene reporter assay. Four plasmids were constructed including two
wildtype with the pGL-3 vector + 3'UTR of their respective genes (HoxB3a or HoxD10a) and
two mutants with the pGL-3 vector + 3'UTR with restriction enzyme site at the seed position
of their respective genes (HoxB3a or HoxD10a). The 3'UTR regions of two genes - HoxB3a
and HoxD10a - from tilapia cDNA were PCR amplified (primers described in Table 1) and
6
individually cloned into the pGL3 vector (Promega) by directional cloning. Fragments were
700bp long (HoxB3a: ENSONIT00000007801 and HoxD10a: ENSONIT00000010838).
Negative controls were constructed mutating the seed region of the miRNA target
gene transcripts as described in Table 2. The mutant constructions were built using the
wildtype plasmids as template. PCR reactions were performed using the primer set consisting
of a primer forward with the desired enzyme site (Table 2) and reverse primer to a specific
region at the plasmid pGL3. To ensure that the mutant plasmids were precisely generated with
the correct nucleotide sequence, we used the Phusion® High-Fidelity DNA Polymerase (New
England BioLab) for the PRC reactions.
2.5. Target validation by luciferase reporter assays
Luciferase reporter assays consist of the following steps. Chick dermal fibroblasts, DF1 cells,
were counted and seeded in 24-well plates (Costar) at 7x103 cells per well and were
maintained for 24 hours in a liquid medium (DMEM with 10% FBS) in the presence of
antibiotic in a CO2 incubator at 37 °C for the perfect adaptation of the plate. Prior to transient
transfection, the medium was replaced with fresh medium without antibiotic. Transfection of
0.4 mg of firefly luciferase reporter vector and 0.02 mg of the control vector containing
Renilla luciferase was performed using lipofectamine 2000 (Invitrogen). Following
transfection DF1 cells were maintained for 24 hours. The next day cells were washed with
PBS and harvested for Dual Luciferase Reporter Assays (Promega) following the
manufacturer's protocol. Each transfection was performed in four wells and repeated three
times independently in different plates. Firefly luciferase activity was normalized to Renilla
luciferase activity.
Assays were quantified using the brightness luciferase. The tests were measured as
follows, wildtype: negative control (plasmid only); plasmid + miRNA of interest; and positive
control (plasmid + miRNA mimic). Mutant test: Negative control (Plasmid-only with the
mutant 3'UTR region); plasmid (mutant 3'UTR region) + miRNA of interest; and positive
control (plasmid with mutant 3'UTR region + miRNA mimic). Renilla luciferase was used as
the control in all the samples.
2.6. In silico analysis of genome organization of miRNA genes
All five miRNAs were analyzed regarding localization and arrangement. For
precursor annotation, mature miRNA sequences were mapped against the Nile tilapia genome
(UCSC, Broad, OreNil1.1) without permitting any mismatches. Retrieved precursor
sequences were both aligned to zebrafish and human homologs and subjected to analysis by
the RNAfold program (http://rna.tbi.univie.ac.at/cgi-bin/RNAfold.cgi) to confirm they fold
into stable stem-looped secondary structures. Then we determined the physical position of
7
pre-miRNAs in the linked groups (LGs) and classified them according to their host region and
strand orientation.
2.7. Statistical Analysis
The data of quantitative PCR were expressed as median ± standard error. The qRT-
PCR and Luciferase gene reporter assay data distribution were parametric then the Two-way
ANOVA test was used. Significant differences were checked by running Bonferroni multiple
comparison tests. Statistical significance was defined as p < 0.05.
3. Results
3.1. Negative correlation between expression profiles of miR-10 family members and Hox
genes
QPCR experiments were sensitive to detect differences between miRNAs and
putative target Hox genes expression signatures throughout distinct developmental stages.
For the miRNAs, the expression of miR-10a-5p, miR-10b-5p, miR-10d-5p, miR-99a-
5p and miR-100-5p generally increased during Nile tilapia embryo development. The profiles
of miR-100-5p and miR-99a-5p were similar, with low expression at 1 dpf and an 8-fold
increase by 3 dpf or 7 dpf respectively (P<0.001). Very similar expression patterns were
observed for miR-10-5p, miR-10b-5p, and miR-10d-5p, which were lowly expressed in 1dpf,
followed by an increase at 3, 5 and 7 dpf and down again at 30 dpf (Figure 2a). By contrast,
the miR-100-5p and miR-99a-5p, gradually increased expression from 3 dpf and remained
highly expressed until 30 dpf (Figure 2a).
For target Hox genes, inversely correlated expression profiles were recovered in
relation to the aforementioned miRNAs. Both HoxA3a HoxB3a and HoxD10a showed high
expression in 1 dpf embryos (P<0.001), followed by decreasing expression during the
subsequent developmental stages with low levels detected at 7 dpf. Interestingly, HoxA3a and
HoxB3a showed a posterior increase in expression at 30 dpf juveniles (P<0.01) (Figure 2b).
Overall, the strikingly contrasting expression signatures of miR-10 and Hox genes
suggested a strong regulatory relationship (Supplementary material – Figure S1).
3.2. Validation of miR-10 family targets in vitro
Luciferase reporter gene analyses were used to confirm the action of miR-10b-5p on
Hox genes. The results confirmed the in silico prediction and expression profiles detected by
qPCR. We observed in DF1 cells transfected with lipofectamine and miR-10b-5p mimics that
both HoxB3a and HoxD10a reactive signal dropped to 50% and 70% of control, respectively,
with no significant change in the mutant constructs (Figures 3a and b). These results are
8
consistent with the qPCR data, and thus reinforcing that miR-10b-5p mediate regulation of
both HoxB3a and HoxD10a.
4. Discussion
4.1. Interaction of miR-10 family members and Hox genes during Nile tilapia development
and maturation
In vertebrates, several miRNAs are known to regulate Hox gene expression (Bender,
2008; Mansfield & McGlinn, 2012). Particularly, the miR-10 miRNA family has a primordial
role in shaping Hox genes expression profiles (Garzon et al., 2006; Woltering & Durston,
2008).
In our analysis, the miR-10 family, as well as Hox genes (HoxA3a, HoxB3a, and
HoxD10a) showed opposite expression patterns in different stages of development. The
miRNAs displayed increased expression in 3, 5 and 7 dpf while Hox genes were decreased in
their expression level. This inverse relationship is important because experimental conditions
have shown that the timing of Hox gene activation produces phenotypic alterations, even in
cases when the final Hox expression patterns are preserved (Zákány et al., 1997; Kondo &
Duboule, 1999). This makes sense when we consider the existence of distinct functional
activities associated with early and late phases of vertebrate Hox gene expression (Carapuço
et al., 2005). It has also been suggested that during early vertebrate development the usually
repressed state of the Hox cluster keeps the late regulatory elements in a "silent state", and
only after global repression is erased these elements become accessible to transcriptional
regulators and, therefore, functional (Tschopp & Duboule, 2011).
At the onset of early development, all three Hox genes were highly expressed (1 dpf)
with HoxA3a exhibiting the highest level of expression from 1 to 3dpf. This finding
reinforces the predominant role of HoxA3a at the early onset of embryonic differentiation and
it is compliant to the collinear expression of body segmentation. HoxA3a is required in
patterning the anterior body plan during embryogenesis process and also for the development
of endodermal, mesodermal, and ectodermal derivatives, cell migration, proliferation,
apoptosis and differentiation, all in mouse (Chen et al., 2010). Moreover the higher
expression of HoxA3a in relation to its paralog HoxB3a here documented match to previous
observations at early segmentation stages of Nile tilapia, where paralog variations in anterior
boundaries of expression were reported (Le Pabic et al., 2009). At the embryonic timing of
3dpf, after the different germ layers are defined and organogenesis begins, neural tube, heart
and somites are formed under high peaks of Hox genes expression and still low expression of
miR-10. Our findings also have shown a remarkable decrease in HoxB3a expression in
embryos from 3dpf to 7dpf, which is in agreement with a rapid fade in HoxB3a expression
9
during organogenesis of pharyngeal segmentation in tilapia (Le Pabic et al., 2009). HoxB3a
was also implicated in the spinal cord formation (Minoux et al., 2009) and was shown to be
repressed by miR-10 in zebrafish (Woltering & Durston, 2008). Both data are consistent with
similar temporal domains of expression detected for HoxB3a co-ortologs between tilapia and
zebrafish (Miller et al., 2000). Our experimental data also suggest that HoxB3a regulation is
constrained in fish since both orthologs from Nile tilapia and zebrafish are regulated by miR-
10b-5p.
Furthermore, we have shown miR-10b-5p can also modulate a second Hox target
gene and from a distinct cluster, the HoxD10a (Figures 4a and b). In fact, the repression of
two or more Hox genes by a single miRNA was previously observed in zebrafish, where miR-
10 represses the nearby HoxB1a and HoxB3a genes and its overexpression also induces the
associated loss of function phenotypes for both (Woltering & Durston, 2008). Also, the
miRNA-mediated regulation of two Hox genes belonging to different clusters was previously
reported in cell lineages in vitro (Chambeyron & Bickmore, 2004; Morey et al., 2007).
However in this work we bring as novelty the detection of miR-10b control over two Hox
genes belonging to distinct clusters in embryos at distinct developmental stages. Therefore,
given the constrained function of Hox clusters in vertebrates, we can presume that both in
Nile tilapia and in zebrafish the formation and specification of fins along the body axis is
dependent of the modulation of HoxD10a expression by the miR-10b-5p.
Another interesting point is that the collinearity of Hox genes expression (spatial,
temporal and quantitative), a common feature of vertebrates (Andrey et al., 2013), implies
that HoxD10a is expressed after other HoxD genes (i.e., HoxD3a, HoxD4a and HoxD9a, in
this order) that are closer to 3' regions. Furthermore, we found that in Nile tilapia the mir-10b
is encoded in the D cluster, upstream HoxD4a gene (Figure 1), a characteristic that can help
its repressive activity over HoxD10a.
A considerable part of metazoan miRNA genes were reported in introns of protein-
coding genes (Rodriguez et al., 2004). miR-10 paralogs in tilapia were not palindromic and,
therefore, cannot be generated from an antisense transcript (Supplementary material - Figure
S2). This strongly suggests that mir-10b paralogs behave as intronic sense-oriented miRNAs
in relation to their encoded Hox genes and are spliced out of the transcript and further
processed into mature miRNAs (Lin et al., 2006). Sense-oriented intronic miRNAs are
thought to be processed as part of their host genes and their expression correlates with that of
their hosts (Bartel, 2004; Berezikov, 2011), although they may also have own promoters
located at intronic upstream regions (Ozsolak et al., 2008). In zebrafish, miR-10 paralogs
associated with Hox4 genes have similar patterns of expression (Woltering & Durston, 2008).
In tilapia, we have accessed miR-10 regulation over Hox3 genes rather than in relation to
their host Hox4 genes. Thus, the negatively correlated expression patterns between miR-10
10
and Hox3 genes makes sense because miR-10 loci are not physically encoded within introns
of HoxA3a, HoxB3a, and HoxD3a, meaning that they would not behave as co-expressed
intronic miRNAs.
4.2. MicroRNAs expression along the development of the Nile tilapia
Overall, the constrained expression pattern component of Hox genes is fundamental
to keep a stable ontogenetic process during embryos development. In this sense, we sought to
compare specific developmental patterning among tilapia and other fishes regarding the
expression of Hox genes and miRNAs, since we have generated profiles that covered basic
early developmental stages of Nile tilapia. Noteworthy, a documented overall slower
embryonic development was reported for tilapia relative to zebrafish, whereas the timing of
formation of diverse body structures is similar in the two species regardless of their respective
sizes (Kimmel et al., 1995; Fujimura and Okada, 2007). In our analysis of miRNAs and target
gene expression profiles from 1, 3 and 5 dpf, we found that all miR-10 family members were
lowly expressed at 1dpf, the time of zygote cleavage, blastula, and early gastrula, precisely
where Hox genes were highly expressed. It is likely that at early developmental stages miR-
10 family low expression may help to keep the primordial high expression of Hox genes and
consequently, the correct development of the embryo. In the comparison of the early
embryonic development of zebrafish and medaka to Nile tilapia, embryos of these three
species were shown to share the same characteristics along to first stages of the
developmental course (from 1 to 5dpf), including formation of the embryonic shield, the
cerebellar primordium, the pectoral fin bud and the aortic arches (Fujimura and Okada, 2007).
From this data, one can speculate, that the miR-10 expression signature seen in Nile tilapia is
probably retained in both zebrafish and medaka species.
Conversely, in the later development, after the transition from the embryonic to the
larval stage, the ontogenesis of Nile tilapia in comparison to the zebrafish and medaka is
slightly accelerated. In Nile tilapia, the transition from embryo to larva at ~7dpf predates yolk
absorption and subsequently the juvenile stage is reached in a shorter time period compared to
these fishes (Fujimura and Okada, 2007). At this stage (7dpf), in which organogenesis was
ended and major body patterning is well defined, the down-regulation of Hox genes was
accompanied by a considerable increase in the expression of miR-10 family members clearly
suggesting a miRNA-mediated inhibition. Therefore in the interval from the larva (7dpf) to
juvenile (30dpf) we see a retake in Hox genes expression and a corresponding decrease in
miR-10b. At the later developmental stages, Hox gene expression may be important for keep
body patterning throughout the growth of the whole body and structures already formed. This
late expression of Hox genes was shown to be important in mammals, where HoxA10 was
expressed in the uterus of healthy adult mice female, with absolute levels ranging along the
11
reproductive cycle. But the absence of HoxA10 expression in the endometrium of these
animals led to infertility due to a failure of embryo implantation (Bagot et al., 2000). Possibly
the Hox gene expression in juvenile and subsequently in adult closely mimics embryonic
expression, but might be involved in cell renewal, as well as, in normal physiological changes
that happen in mature life (Morgan, 2006).
5. Conclusion
In conclusion, miR-10 family members have shown to be important during Nile
tilapia initial development and seem to have a direct effect in modulating the expression of
Hox genes. The balance between expression of miRNAs and Hox genes is coordinated as in
an orchestra, where each element must have its peak at a specific time and the interaction
between its modulations that leads to the perfect compass of development (Tanzer et al.,
2005). In Nile tilapia genetic orchestra, the miR-10b-5p proved to be a potent regulator,
coordinating HoxB3a and HoxD10a expression (Figure 4), genes that in turn known to
encode DNA binding proteins that specify cells in the spinal cord and in the development of
members of vertebrate embryos. Also, we experimentally evidenced that miR-10 may
regulate not only Hox genes co-located at the same cluster but also paralogs from other Hox
gene clusters. For instance, the miR-10b precursor is encoded in the HoxD but was
demonstrated to target a paralog gene at HoxB.
Another important element refers to the genomic localization of miRNAs and target
genes. The physical proximity between the elements facilitates their interaction allowing the
control the primordial body segmentation in a chronological order during Nile tilapia early
developmental process, assuring the perfect formation of the organism. Furthermore, the
evolutionary constrained processes appear to underlie the complexity of the miRNA-mediated
regulation of Hox genes mechanisms, which includes multiple global and local transcriptional
elements. Overall, our results contribute to clarify the molecular pathways underlying early
development in teleosts, especially in Nile tilapia fish.
Acknowledgements
We would like to thanks Sao Paulo Research Foundation (FAPESP) and Higher
Education Personnel Training Coordination (CAPES) for financial support, Dr Alexandre
Wagner Silva Hilsdorf and Dr David Penman and Royal Fish Farm for providing Nile tilapia
embryos, and Camille Viaut for the support with the luciferase essays.
12
References
Alexander, T., Nolte, C., Krumlauf, R. 2009. Hox genes and segmentation of the hindbrain
and axial skeleton. Ann. Rev. Cell. Dev. Biol. 25, 431–56.
Amores, A., Suzuki, T., Yan, Y.L., Pomeroy, J., Singer, A., Amemiya, C., Postlethwait, J.H.,
2004. Developmental roles of pufferfish Hox clusters and genome evolution in ray-fin
fish. Genome Res. 14, 1–10.
Andrey, G., Montavon, T., Mascrez, B., Gonzalez, F., Noordermeer, D., Leleu, M., 2013. A
switch between topological domains underlies HoxD genes collinearity in mouse limbs.
Science 340, 6137.
Bagot, C.N., Troy, P.J., Taylor, H.S., 2000. Alteration of maternal Hoxa10 expression by in
vivo gene transfection affects implantation. Gene Ther. 7, 1378-1384.
Bartel, D.P., 2009. MicroRNAs: target recognition and regulatory functions. Cell 136, 215-
233.
Bender, W., 2008. MicroRNAs in the Drosophila bithorax complex. Genes Dev. 22, 14-19.
Berezikov, E., 2000. Evolution of microRNA diversity and regulation in animals. Nat. Rev.
Genet. 12, 846-860.
Carapuço, M., Nóvoa, A., Bobola, N., Mallo, M., 2005. Hox genes specify vertebral types in
the presomitic mesoderm. Genes Dev. 19, 2116-2121.
Chambeyron, S., Bickmore, W.A., 2004. Chromatin decondensation and nuclear
reorganization of the HoxB locus upon induction of transcription. Genes Dev. 10, 1119-
30.
Chen, L., Zhao, P., Wells, L., Amemiya, C.T., Condie, B.G., Manley, N.R., 2010. Mouse and
zebrafish Hoxa3 orthologues have nonequivalent in vivo protein function. Proc. Natl.
Acad. Sci. USA. 107, 10555–10560.
Denell, R.E., Hummels, K.R., Wakimoto, B.T., Kaufman, T.C., 1981. Developmental studies
of lethality associated with the antennapedia gene complex in Drosophila melanogaster.
Dev Biol. 81, 43–50.
Duboule, D., 2007. The rise and fall of Hox gene clusters. Development 134, 2549–60.
Enright, A.J., John, B., Gaul, U., Tuschl, T., Sander, C., Marks, D.S., 2003. miRanda
algorithm: MicroRNA targets in Drosophila. Genome Biol. 5,R1.
Fujimura, K., Okada, N., 2007. Development of the embryo, larva and early juvenile of Nile
tilapia Oreochromis niloticus (Pisces: Cichlidae), developmental staging system. Dev.
Growth Differ. 49,301-24.
Garzon, R., Pichiorri, F., Liu, C.G., Bloomfield, C.D., Andreeff, M., Croce, C.M., 2006.
MicroRNA fingerprints during human mega karyocytopoiesis. Proc. Natl. Acad. Sci. USA.
103, 5078–5083.
13
Gellon, G., McGinnis,W., 1998. Shaping animal body plans in development and evolution by
modulation of Hox expression patterns. BioEssays 20, 116–125.
Gendron-Maguire, M., Mallo, M., Zhang, M., Gridley, T., 1993. Hoxa-2 mutant mice exhibit
homeotic transformation of skeletal elements derived from cranial neural crest. Cell 75,
1317–31.
Griffiths-Jones, S., 2004. The microRNA registry. Nucleic Acids Res. 32,D109–D111.
Grimson, A., Farh, K.K., Johnston, W.K., Garrett-Engele, P., Lim, L.P., Bartel, D.P. 2007.
MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol. Cell
27, 91-105.
Ijiri, S., Kaneko, H., Kobayashi, T., Wang, D., Sakai, F., Paul-Prasanth, B., Nakamura, M.,
Nagahama, Y., 2008. Sexual dimorphic expression of genes in gonads during early
differentiation of a teleost fish, the Nile tilapia Oreochromis niloticus. Biol.Reprod. 78,
333–341.
Kaufman ,T.C., 1978. Cytogenetic analysis of chromosome 3 in Drosophila melanogaster:
isolation and characterization of four new alleles of the proboscipedia (pb) locus. Genetics
90, 579–96.
Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T.F., 1995. Stages of
embryonic development of the zebrafish. Dev. Dyn. 203, 2553-2563.
King, B.L., Gillis, J.A.,Carlisle, H.R., Dahn, R.D., 2011. A natural deletion of the HoxC
cluster in elasmobranch fishes. Science 1517, 334-350.
Kondo, T., Duboule, D., 1999. Breaking colinearity in the mouse HoxD complex. Cell 97,
407-417.
Krek, A., Grün, D., Poy, M.N., Wolf , R., Rosenberg, L., Epstein, E.J., MacMenamin, P.,
Gunsalus, K.C., Rajewsky, M., 2005. Combinatorial microRNA target predictions. Nat.
Gen. 37, 495-500.
Le Pabic, P., Stellwag, E.J., Scemama, J.L., 2009. Embryonic development and
skeletogenesis of the pharyngeal jaw apparatus in the cichlid Nile Tilapia (Oreochromis
niloticus). Anat. Rec. 292, 1780-1800.
Lee, D., Shin, C., 2012. MicroRNA-target interactions: new insights from genome-wide
approaches. Ann. N. Y. Acad. Sci. 1271, 118-128.
Lin, S.L., Miller, J.D., Ying, S.Y., 2006. Intronic MicroRNA (miRNA). J. Biomed.
Biotechnol. 6, 26818
Lung AH (2010). miR-10 in development and cancer. Cell Death Differ. 17, 209–214.
Mallo, M., Alonso, C.R., 2013. The regulation of Hox gene expression during animal
development. Development 140, 3951–63.
Mansfield, J.H., McGlinn, E., 2012. Evolution, expression, and developmental function of
Hox-embedded miRNAs. Curr. Top. Dev. Biol. 99, 31-57.
14
Miller, C.T., Schilling, T.F., Lee, K.H., Parker, J., Kimmel, C.B., 2000. Sucker encodes a
zebrafish Endothelin-1 required for ventral pharyngeal arch development. Development
127, 3815–3828.
Minoux, M., Antonarakis, G.S., Kmita, M., Duboule, D., Rijli, F.M., 2009. Rostral and caudal
pharyngeal arches share a common neural crest ground pattern. Development 136, 637–
645.
Montavon, T., Soshnikova, N., Mascrez, B., Joye, E., Thevenet, L., Splinter, E., De Laat, W.,
Spitz, F., Duboule, D., 2011. A regulatory archipelago controls Hox genes transcription in
digits. Cell 147, 1132-1145.
Morey, C., Da Silva, N.R., Perry, P., Bickmore, W.A., 2007. Nuclear reorganization and
chromatin decondensation are conserved, but distinct, mechanisms linked to Hox gene
activation. Development. 134, 909-919.
Morgan, R., 2006. Hox genes: a continuation of embryonic patterning?. Trends Genet. 22, 67-
69.
Ozsolak, F., Poling, L.L., Wang, Z., Liu, H., Liu, X.S., Roeder, R.G., Zhang, X., Song, J.,
Fisher, D.E., 2008. Chromatin structure analyses identify miRNA promoters. Genes and
Development 22, 3172-3183.
Rijli, F.M., Mark, M., Lakkaraju , S., Dierich, A., Dolle, P., Chambon, P., 1993. A homeotic
transformation is generated in the rostral branchial region of the head by disruption of
Hoxa-2, which acts as a selector gene. Cell 75, 1333–1382.
Rodriguez, A., Griffiths-Jones, S., Ashurst, J.L., Bradley, A., 2004. Identification of
mammalian microRNA host genes and transcription units. Genome Res. 14, 1902–1910.
Santini, S., Bernardi, G., 2005. Organization and base composition of tilapia Hox genes:
implications for the evolution of Hox clusters in fish. Gene 346, 51–61.
Tanzer, A., Amemiya, C.T., Kim, C.B., Stadler, P.F., 2005. Evolution of microRNAs located
within Hox gene clusters. J Exp Zool B Mol Dev Evol. 304, 75-85.
Tehler, D., Høyland-Kroghsbo, N.M., Lund, A.H., 2011. The miR-10 microRNA precursor
family. RNA Biol. 8, 728-734.
Tschopp, P., Duboule, D., 2011. A genetic approach to the transcriptional regulation of Hox
gene clusters. Annu. Rev. Genet. 45,145-166.
Van de Ven, C., Bialecka, M., Neijts, R., Young, T., Rowland, J.E., Stringer, E.J., 2011.
Concerted involvement of Cdx/Hox genes and Wnt signaling in morphogenesis of the
caudal neural tube and cloacal derivatives from the posterior growth zone. Development
138, 3451–62.
Zákány, J., Gérard, M., Favier, B., Duboule, D., 1997. Deletion of a HoxD enhancer induces
transcriptional heterochrony leading to transposition of thesacrum. EMBO J. 16, 4393-
4402.
15
Zákány, J., Duboule, D., 2007. The role of Hox genes during vertebrate limb development.
Curr. Opin. Genet. Dev. 17, 359–66.
Woltering, J.M., Durston, A.J., 2008. MiR-10 Represses HoxB1a and HoxB3a in zebrafish.
PLoS One 1, e1396.
Wong, S.F.L., Agarwal, V., Mansfield, J.H., Denansh, N., Schwartzf, M.G., Prosseri, H.M.,
Pourquié, O., Bartel, D.P., Tabin, C.J., McGlinn, E., 2015. Independent regulation of
vertebral number and vertebral identity by microRNA-196 paralogs. Proc. Natl. Acad. Sci.
USA 112, 4884–4893.
16
Tables
Table 1. MicroRNA assays and Hox genes primers used for qPCR analyzes
miRNA ID Mature miRNA Sequence
hsa-miR-10a-5p ACCCUGUAGAUCCGAAUUUGU
hsa-miR-10b-5p UACCCUGUAGAACCGAAUUUGU
dre-miR-10d-5p UACCCUGUAGAACCGAAUGUGUG
hsa-miR-99a-5p AACCCGUAGAUCCGAUCUUGUG
hsa-miR-100-5p AACCCGUAGAUCCGAACUUGUG
RNU6B CGCAAGGAUGACACGCAAAUUCGUGAAGCGUUCCAUAUUUUU
mRNA Primer Sequence
HoxA3a Forward: TAACCGAACGGCAGGTGAAA
Reverse: TGTCGCTGGATTCATGGCTT
HoxB3a Forward: TCTGGAAGCCGTTTTCCTCC
Reverse: ACGTGACGGTGTCTTTCCAA
HoxD10a Forward: CTGAATCGTGTCCGGTCGAT
Reverse: TGCTTCCCGTTCGCATAA
HPRT Forward: GACATCATGGATGACATGGGGG
Reverse: GTAGTCGAGCAGGTCTGCAAAAA
Table 2. Mutant seed region created by SmaI restriction site insertion (italics underlined) in
the Hox genes investigated.
Gene Seed sequence in 3'UTR region
HoxA3a TTACGGCATCGGGGTGCTCTGACCCCGGGGCGGCTCGTCACTCTCCGCCGGCGTTGG
HoxB3a ACGGCATCGGGGTGCTCTGTCCCCGGGGCGCCTCGTCACTCTCCGCCGG
HoxD10a GCGTCCTGTATTTGTTTTGTTTCATCCCCGGGTAACACTTTGTGGTCTTAAATTATT
17
Figures
Figure 1. Hox gene clusters and their organization in Oreochromis niloticus (green boxes)
and Danio rerio (blue boxes). MicroRNAs encoded by the Hox cluster are indicated and red
lines in O. niloticus represent the components of the Hox gene family regulated by miRNA.
Dark blue squares in D. rerio represent pseudogenes.
18
Figure 2. Relative quantification of microRNA (a) and mRNA transcripts (b) expression
during Nile tilapia development based on qRT-PCR (log10). Relative expression was
normalized against U6 snRNA to miRNAs and HPRT to mRNA. ***P<0.001, **P<0.01 and
*P<0.05 as assessed by Two-way ANOVA test with Bonferroni correction.
19
Figure 3. Luciferase assay with the Pgl-3 vector with the 3'UTR of HoxB3a (a) and HoxD10a
(b) under the action of miR-10b-5p (a and b). The first section demonstrates the relative level
of luciferase activity after transfection of DF-1 cells with 3'UTR region wild only (Negative
control); 3'UTR wild+mimic control (Positive control) and 3'UTR wild+miR mimic. Second
section: 3'UTR mutant only (Negative control); 3'UTR mutant+mimic control (Positive
control) and 3'UTR mutant+miR mimic. Bars represent the normalized average of relative
luciferase units. *P<0.05 as assessed by Two-way ANOVA test with Bonferroni correction.
20
Figure 4. Synthetic view of miRNA action in the control of Hox genes expression in Nile
tilapia (a). The detailed alignment between the miRNA and its target mRNA (b) can be
observed. The hairpin structure of pre-miR-10b-5p is highlighted in black and the mature
sequences 3p and 5p of the miRNA are highlighted in blue (c).
21